We now support re-running tests in batch on Openlayer, so that you can recalculate results when new data is added retroactively. In addition, we’ve added a ton of new integrations – Ruby and Strands Agents SDKs, Gemini API and Copilot Studio integrations, and more – so that Openlayer fits in seamlessly with the workflows you’re already running. As always, we’ve made continuous improvements to the UI and docs, so your experience will be even smoother. Read the full list below.
Features
•PlatformAutomatic threshold support for tests based on custom metrics
•SDKsOpenlayer Ruby SDK
•PlatformCustom time range selection for test execution
•PlatformManual test re-runs
•PlatformNew primary key uniqueness constraint test
•IntegrationsStrands Agents SDK integration
•IntegrationsMicrosoft Copilot Studio integration
•IntegrationsGoogle Gemini API integration
•APIAPI endpoints for retrieving test results and managing projects
Improvements
•UI/UXClearer trace visualizations
•PlatformMore informative skipped test messages
•PlatformMore reliable OpenTelemetry support
•UI/UXSmoother Snowflake connection flow with support for views
•PlatformSimplified Databricks onboarding flow
•PlatformMore robust directory sync configuration
•PlatformBetter session and user tracking for GenAI projects
•UI/UXRedesigned monitoring test results graph
•PlatformCharacteristic Stability Index (CSI) as a drift detection method
Fixes
•UI/UXEvidence type selection fixed when creating governance rules
•PlatformUser email matching is no longer case-sensitive
•SDKsStep detection in LangChain integration
•SDKsModel output capture in Google ADK integration
•SDKsDuplicate trace steps in LiteLLM tracer
$ openlayer push
Stop guessing.
Ship with confidence.
The AI governance and observability platform
We value your privacy
We use cookies to enhance your browsing experience, serve personalized content, and analyze our traffic.
